This skill should be used when the user asks to "add MCP App support to my web app", "turn my web app into a hybrid MCP App", "make my web page work as an MCP App too", "wrap my existing UI as an MCP App", "convert iframe embed to MCP App", "turn my SPA into an MCP App", or needs to add MCP App support to an existing…
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends an AI agent's capabilities with specialized knowledge, workflows, or tool integrations.
Migrate a project to a newer version of the tldraw SDK. Use when upgrading tldraw packages, fixing TypeScript errors after a tldraw upgrade, or when the user mentions tldraw migration.
A skill that uses the Gemini website, Google's AI assistant, for conversations and image generation. It defines how the assistant should choose tools and handle long-running requests.
Manual-invocation only. Five-phase feature loop (audit → ground → fixture → implement → verify) for driving a single SVG feature to Chromium parity in the grida htmlcss::svg renderer. Sibling to dev-render-htmlcss-feature (HTML/CSS path); same loop shape, different corpus (resvg-test-suite + Chrome bake) and different…
Guides performance optimization work on the Grida Canvas Rust rendering engine (grida crate). Use when asked to benchmark, profile, optimize, or analyze rendering performance — panning, zooming, compositing, caching, culling, or frame budgeting.
Design, review, and debug visual rendering tests: reftests (independent oracle), golden/snapshot regression tests, probe tests (pixel assertions without vision), SVG reftest suites, and Figma refig suites. Use when adding pixel diffs, choosing an oracle strategy, or comparing renderer output against a ground truth.
Use when doing placement and metal routing (manual or automatic) on GDS layouts via the klayout-draw-mcp MCP server — standard-cell rows, multi-layer wires, vias, and obstacle-aware maze (Lee) routing, with DRC checking in the loop.
Ask the user to draw or annotate something and get it back as an image. Use when the user runs /draw, says they want to sketch or diagram, or when words alone are failing to pin down a visual change and it would be faster for them to point at it.